Neuroanatomical abnormalities in schizophrenia may be attributable to genetic factors, environmental factors, or their interaction. Thus, neuroimaging studies of sibling pairs that vary in clinical, neuromorphological and genetic factors provide a powerful approach to evaluating brain structure abnormalities associated with schizophrenia. If specific genes underlie neurodevelopmental abnormalities involved in the pathogenesis of schizophrenia, such genes could be discovered by examining the degree to which they segregate with brain structure abnormalities in schizophrenia probands and their siblings. Recently, we have developed computational tools for the analysis of brain structure that permit the detailed analysis of brain structure volumes and shapes. We have used these to compare the structure of the hippocampus and thalamus in schizophrenia subjects and matched controls, and in these studies, shape deformities were found to help discriminate between groups. In addition to these structures, we have now begun the assessement of specific regions of the cerebral cortex (i.e., cingulate gyrus, superior, middle and inferior frontal gyrus, and superior lingual gyrus). Because of their precision, computational tools for brain structure analysis should be an ideal method for discovering subtle abnormalities of brain structure in the nonpsychotic siblings of schizophrenia subjects. The specific aims of this Project are: Aim 1. To collect high resolution MR scans from 50 schizophrenia probands and 50 of their non-psychotic siblings who are within the age of risk for developing the disorder, and from 50 control subjects and 50 of their siblings; Aim 2. To test the hypothesis that measures of brain structure volume, thickness, shape and symmetry can discriminate groups of schizophrenia probands, the siblings of schizophrenia probands, control subjects and the siblings of control subjects; Aim 3. To estimate the heritability of the selected brain structure measures in both schizophrenia and control sibling pairs; Aim 4. To collect longitudinal data in all subjects, and to preliminarily test the hypothesis that neuromorphometric measures can be used to discriminate between siblings of schizophrenia probands that do and do not develop psychopathology during the period of study.